From an idea to working together

Startup requirements

An idea and the drive to build it.

You understand the problem and are ready to build a solution. We help with what is missing for the next step.

A concept, prototype or MVP. On your own or with a team.

  1. Customer pain
  2. Feasible solution
  3. Your contribution

A project foundation

+ Intecrator support

People, experience and resources for an agreed plan

From an assumption to action

An idea proposes. A project tests.

A project starts with a specific task and the first steps towards solving it, not with company registration.

Who is it for?
Still an idea

Small businesses.

A project foundation

Sales staff at small wholesalers who manually transfer orders from emails.

What is the problem?
Still an idea

They have lots of routine work.

A project foundation

Products get mixed up in orders; staff have to check everything again, delaying dispatch.

How do you know?
Still an idea

I think they need it.

A project foundation

I have worked with this process, examined examples and spoken to the people doing it.

What comes first?
Still an idea

AI will automate the entire company.

A project foundation

A tool drafts an order from an email for a staff member to check before saving.

Who does what?
Still an idea

I need a team to do everything.

A project foundation

I document the process, recruit people for testing and evaluate results; I need development and integration support.

A hypothetical wholesale example. You may have no code yet. Observations, conversations and an investigated workflow already provide a basis for work.

Three foundations

What needs to come together

Someone feels the pain

A specific person is losing time, money or opportunities.

  • Identify the user, their task and the situation where the problem occurs.
  • Show how they handle it today and what is unsatisfactory.
  • Distinguish the user from the buyer: the person using the product may not pay for it.

What happened the last time the customer faced this problem?

“Interesting idea” does not establish demand. A real incident, access to examples and willingness to test reveal more. There is no fixed interview quota.

It can be built

There is a feasible first workflow and a way to test the main risk.

  • Describe the input and the useful result the customer receives.
  • Check access to technology, data and integrations. If missing, explain how you could obtain it.
  • Estimate the main costs and whether the benefit justifies them for the buyer.

What is the smallest test that will show the solution works?

For AI, explain how to detect errors and where human review is needed. For a scientific hypothesis, identify an experiment. There is no universal accuracy or revenue threshold.

You can move it forward

The founder has a personal contribution and takes responsibility for the next step.

  • Show experience, research, code, a process description or access to prospective users.
  • Explain your own role, the time you can commit and the expertise you lack.
  • Be ready to talk to customers, test assumptions and change the solution.

What part of this project depends specifically on you?

You need not be a programmer, salesperson and manager at once. Domain expertise is a contribution too. Previous company-building experience is not required.

What this looks like in practice

One task. Your contribution. Our help.

Three hypothetical examples, not participant stories or a guarantee of admission.

Documents in wholesale trade

Initial idea

«Build AI for document processing»

First useful workflow

The founder knows the sales department's work. They have investigated how staff copy orders from emails, where mistakes occur and who corrects them. The first workflow is to draft an order and flag ambiguous items for human review.

The founder takes on

Document the process, obtain examples permitted for testing and arrange a test with sales staff.

Help needed

Development, integration with the business system, and evaluation of errors and processing costs.

What to test

The first test should establish whether total work actually decreases once draft checking is included.

Enquiries to a service company

Initial idea

«A bot that sells instead of a sales representative»

First useful workflow

A developer has spoken to service companies and found that enquiries stall because details are missing: equipment type, the fault and a suitable appointment time. The first workflow is to collect these details and pass a structured request to a dispatcher.

The founder takes on

Build a prototype and test the workflow with users.

Help needed

Find a pilot company, connect the CRM and define a payment model.

What to test

The test should reveal whether requests become more complete and whether customers abandon the conversation. The existence of a bot does not prove this.

Quality control in manufacturing

Initial idea

«A neural network will detect any defect»

First useful workflow

A researcher has an image analysis method and chooses one product type and one class of defects. They investigate the current inspection process, whether representative images can be obtained and the conditions needed for the method to work on a production line.

The founder takes on

Test the method, describe its limits and analyse errors.

Help needed

Access to a pilot site, engineering integration and an economic assessment.

What to test

If data access is still missing, that is the next issue to investigate, not a reason to promise a finished system.

You do not need an established company

Bring the foundation. We develop the rest together.

A finished MVP, revenue, a complete team or a perfect presentation are not universal prerequisites. The scope of support depends on the stage and an agreed plan.

For our first conversation

  • A specific problem and the people it affects.
  • Your work so far, with facts and assumptions clearly separated.
  • Personal involvement and a clear request for help.

We can develop together

  • Demand validation, a prototype and a development plan.
  • Development capacity, legal and operational support.
  • Pilots, sales and preparation for investment.

We work with early B2B and B2B2C technology projects that can scale internationally. AI is a priority, not a requirement. We will also discuss roles, time, rights to existing work and data, and English among key participants.

If you only have a broad idea, start by talking to a specific user. This format will not fit if you expect someone else to build and run the business without your involvement.

Start with clarity

What to put in your application

Use your own words. No special presentation or complex terminology needed.

  1. Who you help

    The user, the buyer and how to reach them.

  2. The problem

    A real situation, its consequences and the current way of working.

  3. Your solution

    One initial workflow and its expected benefit.

  4. What you know

    Observations, interview or test results; identify untested assumptions separately.

  5. Why you can do it

    Your experience, skills, existing work and participants' roles.

  6. What is holding you back

    A specific gap in expertise, people, access or resources.

  7. The next step

    What you want to test with our help and how you will assess the result.

  8. Applications are open throughout the year.

Attach a demo or presentation if available. Do not send confidential customer data: describe the process with an anonymised example you are entitled to use.

What happens next

  1. Application
  2. Conversation and work review
  3. Hypothesis test if needed
  4. Decision on collaboration

The thinking behind the approach

Practical reading on testing ideas, customer needs and solution feasibility.

Apply to Intecrator

Three short steps. Applications are open all year.

Project
Contacts
Materials

For investors

Leave your contacts and we’ll get in touch.

Send a message

Thank you!

We have received your message and will write to you by e-mail.
Apply